commit ghc-statistics for openSUSE:Factory
Script 'mail_helper' called by obssrc Hello community, here is the log from the commit of package ghc-statistics for openSUSE:Factory checked in at 2022-08-01 21:30:34 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Comparing /work/SRC/openSUSE:Factory/ghc-statistics (Old) and /work/SRC/openSUSE:Factory/.ghc-statistics.new.1533 (New) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Package is "ghc-statistics" Mon Aug 1 21:30:34 2022 rev:5 rq:987093 version:0.16.1.0 Changes: -------- --- /work/SRC/openSUSE:Factory/ghc-statistics/ghc-statistics.changes 2022-02-11 23:11:39.851337240 +0100 +++ /work/SRC/openSUSE:Factory/.ghc-statistics.new.1533/ghc-statistics.changes 2022-08-01 21:30:59.809759381 +0200 @@ -1,0 +2,18 @@ +Thu May 12 18:58:41 UTC 2022 - Peter Simons <psimons@suse.com> + +- Update statistics to version 0.16.1.0. + Upstream has edited the change log file since the last release in + a non-trivial way, i.e. they did more than just add a new entry + at the top. You can review the file at: + http://hackage.haskell.org/package/statistics-0.16.1.0/src/changelog.md + +------------------------------------------------------------------- +Fri Mar 18 18:26:26 UTC 2022 - Peter Simons <psimons@suse.com> + +- Update statistics to version 0.16.0.2. + ## Changes in 0.16.0.2 + + * Bug in constructor of binomial distribution is fixed (#181). It accepted + out-of range probability before. + +------------------------------------------------------------------- Old: ---- statistics-0.16.0.1.tar.gz New: ---- statistics-0.16.1.0.tar.gz ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Other differences: ------------------ ++++++ ghc-statistics.spec ++++++ --- /var/tmp/diff_new_pack.Ssj58q/_old 2022-08-01 21:31:00.437761182 +0200 +++ /var/tmp/diff_new_pack.Ssj58q/_new 2022-08-01 21:31:00.441761194 +0200 @@ -19,7 +19,7 @@ %global pkg_name statistics %bcond_with tests Name: ghc-%{pkg_name} -Version: 0.16.0.1 +Version: 0.16.1.0 Release: 0 Summary: A library of statistical types, data, and functions License: BSD-2-Clause @@ -33,8 +33,8 @@ BuildRequires: ghc-deepseq-devel BuildRequires: ghc-dense-linear-algebra-devel BuildRequires: ghc-math-functions-devel -BuildRequires: ghc-monad-par-devel BuildRequires: ghc-mwc-random-devel +BuildRequires: ghc-parallel-devel BuildRequires: ghc-primitive-devel BuildRequires: ghc-random-devel BuildRequires: ghc-rpm-macros ++++++ statistics-0.16.0.1.tar.gz -> statistics-0.16.1.0.tar.gz ++++++ diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Correlation.hs new/statistics-0.16.1.0/Statistics/Correlation.hs --- old/statistics-0.16.0.1/Statistics/Correlation.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Correlation.hs 2001-09-09 03:46:40.000000000 +0200 @@ -30,7 +30,7 @@ pearson = correlation {-# INLINE pearson #-} --- | Compute pairwise pearson correlation between rows of a matrix +-- | Compute pairwise Pearson correlation between rows of a matrix pearsonMatByRow :: Matrix -> Matrix pearsonMatByRow m = generateSym (rows m) @@ -43,7 +43,7 @@ -- Spearman ---------------------------------------------------------------- --- | compute spearman correlation between two samples +-- | compute Spearman correlation between two samples spearman :: ( Ord a , Ord b , G.Vector v a @@ -64,7 +64,7 @@ (x, y) = G.unzip xy {-# INLINE spearman #-} --- | compute pairwise spearman correlation between rows of a matrix +-- | compute pairwise Spearman correlation between rows of a matrix spearmanMatByRow :: Matrix -> Matrix spearmanMatByRow = pearsonMatByRow . fromRows . fmap rankUnsorted . toRows diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Distribution/Binomial.hs new/statistics-0.16.1.0/Statistics/Distribution/Binomial.hs --- old/statistics-0.16.0.1/Statistics/Distribution/Binomial.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Distribution/Binomial.hs 2001-09-09 03:46:40.000000000 +0200 @@ -165,7 +165,7 @@ -> Maybe BinomialDistribution binomialE n p | n < 0 = Nothing - | p >= 0 || p <= 1 = Just (BD n p) + | p >= 0 && p <= 1 = Just (BD n p) | otherwise = Nothing errMsg :: Int -> Double -> String diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Distribution/Exponential.hs new/statistics-0.16.1.0/Statistics/Distribution/Exponential.hs --- old/statistics-0.16.0.1/Statistics/Distribution/Exponential.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Distribution/Exponential.hs 2001-09-09 03:46:40.000000000 +0200 @@ -11,7 +11,7 @@ -- Portability : portable -- -- The exponential distribution. This is the continuous probability --- distribution of the times between events in a poisson process, in +-- distribution of the times between events in a Poisson process, in -- which events occur continuously and independently at a constant -- average rate. diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Distribution/Poisson/Internal.hs new/statistics-0.16.1.0/Statistics/Distribution/Poisson/Internal.hs --- old/statistics-0.16.0.1/Statistics/Distribution/Poisson/Internal.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Distribution/Poisson/Internal.hs 2001-09-09 03:46:40.000000000 +0200 @@ -61,7 +61,7 @@ 1.4189385332046727 + 0.5 * log lambda + zipCoefficients lambda coefficients --- | Returns the average of the upper and lower bounds accounding to +-- | Returns the average of the upper and lower bounds according to -- theorem 2. alyThm2 :: Double -> [Double] -> [Double] -> Double alyThm2 lambda upper lower = @@ -164,7 +164,7 @@ dropWhile (not . (< negate m_epsilon * lambda)) $ [ let x = probability lambda k in x * log x | k <- [0..]] --- | Compute the entropy of a poisson distribution using the best available +-- | Compute the entropy of a Poisson distribution using the best available -- method. poissonEntropy :: Double -> Double poissonEntropy lambda diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Distribution/Weibull.hs new/statistics-0.16.1.0/Statistics/Distribution/Weibull.hs --- old/statistics-0.16.0.1/Statistics/Distribution/Weibull.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Distribution/Weibull.hs 2001-09-09 03:46:40.000000000 +0200 @@ -10,7 +10,7 @@ -- Stability : experimental -- Portability : portable -- --- The weibull distribution. This is a continuous probability +-- The Weibull distribution. This is a continuous probability -- distribution that describes the occurrence of a single event whose -- probability changes over time, controlled by the shape parameter. @@ -38,7 +38,7 @@ import Statistics.Internal --- | The weibull distribution. +-- | The Weibull distribution. data WeibullDistribution = WD { wdShape :: {-# UNPACK #-} !Double , wdLambda :: {-# UNPACK #-} !Double @@ -99,11 +99,11 @@ instance D.ContGen WeibullDistribution where genContVar d = D.genContinuous d --- | Standard weibull distribution with scale factor (lambda) 1. +-- | Standard Weibull distribution with scale factor (lambda) 1. weibullStandard :: Double -> WeibullDistribution weibullStandard k = weibullDistr k 1.0 --- | Create weibull distribution from parameters. +-- | Create Weibull distribution from parameters. -- -- If the shape (first) parameter is @1.0@, the distribution is equivalent to a -- 'Statistics.Distribution.Exponential.ExponentialDistribution' with parameter @@ -114,7 +114,7 @@ -> WeibullDistribution weibullDistr k l = either error id $ weibullDistrErr k l --- | Create weibull distribution from parameters. +-- | Create Weibull distribution from parameters. -- -- If the shape (first) parameter is @1.0@, the distribution is equivalent to a -- 'Statistics.Distribution.Exponential.ExponentialDistribution' with parameter @@ -134,7 +134,7 @@ ++ " and lambda " ++ show l --- | Create weibull distribution from mean and standard deviation. +-- | Create Weibull distribution from mean and standard deviation. -- -- The algorithm is from "Methods for Estimating Wind Speed Frequency -- Distributions", C. G. Justus, W. R. Hargreaves, A. Mikhail, D. Graber, 1977. diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Distribution.hs new/statistics-0.16.1.0/Statistics/Distribution.hs --- old/statistics-0.16.0.1/Statistics/Distribution.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Distribution.hs 2001-09-09 03:46:40.000000000 +0200 @@ -47,7 +47,7 @@ class Distribution d where -- | Cumulative distribution function. The probability that a -- random variable /X/ is less or equal than /x/, - -- i.e. P(/X/≤/x/). Cumulative should be defined for + -- i.e. P(/X/���/x/). Cumulative should be defined for -- infinities as well: -- -- > cumulative d +��� = 1 @@ -84,14 +84,14 @@ class Distribution d => ContDistr d where -- | Probability density function. Probability that random -- variable /X/ lies in the infinitesimal interval - -- [/x/,/x+/δ/x/) equal to /density(x)/⋅δ/x/ + -- [/x/,/x+/��/x/) equal to /density(x)/�����/x/ density :: d -> Double -> Double density d = exp . logDensity d -- | Natural logarithm of density. logDensity :: d -> Double -> Double logDensity d = log . density d -- | Inverse of the cumulative distribution function. The value - -- /x/ for which P(/X/≤/x/) = /p/. If probability is outside + -- /x/ for which P(/X/���/x/) = /p/. If probability is outside -- of [0,1] range function should call 'error' quantile :: d -> Double -> Double quantile d x = complQuantile d (1 - x) diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Resampling/Bootstrap.hs new/statistics-0.16.1.0/Statistics/Resampling/Bootstrap.hs --- old/statistics-0.16.0.1/Statistics/Resampling/Bootstrap.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Resampling/Bootstrap.hs 2001-09-09 03:46:40.000000000 +0200 @@ -1,4 +1,3 @@ -{-# LANGUAGE CPP #-} -- | -- Module : Statistics.Resampling.Bootstrap -- Copyright : (c) 2009, 2011 Bryan O'Sullivan @@ -31,9 +30,7 @@ import qualified Statistics.Resampling as R -#if !defined(__GHCJS__) -import Control.Monad.Par (parMap, runPar) -#endif +import Control.Parallel.Strategies (parMap, rdeepseq) data T = {-# UNPACK #-} !Double :< {-# UNPACK #-} !Double infixl 2 :< @@ -51,15 +48,7 @@ -- this. -> [Estimate ConfInt Double] bootstrapBCA confidenceLevel sample resampledData -#if defined(__GHCJS__) - -- monad-par causes seems to cause "thread blocked indefinitely on MVar" - -- on GHCJS still - -- - -- I (phadej) would change the interface to return IO, and use mapConcurrently from async - = map e resampledData -#else - = runPar $ parMap e resampledData -#endif + = parMap rdeepseq e resampledData where e (est, Bootstrap pt resample) | U.length sample == 1 || isInfinite bias = diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/Statistics/Sample.hs new/statistics-0.16.1.0/Statistics/Sample.hs --- old/statistics-0.16.0.1/Statistics/Sample.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/Statistics/Sample.hs 2001-09-09 03:46:40.000000000 +0200 @@ -51,7 +51,7 @@ , fastVarianceUnbiased , fastStdDev - -- * Joint distirbutions + -- * Joint distributions , covariance , correlation , pair @@ -215,7 +215,7 @@ -- $variance -- --- The variance—and hence the standard deviation—of a +-- The variance ��� and hence the standard deviation ��� of a -- sample of fewer than two elements are both defined to be zero. -- $robust diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/benchmark/bench.hs new/statistics-0.16.1.0/benchmark/bench.hs --- old/statistics-0.16.0.1/benchmark/bench.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/benchmark/bench.hs 2001-09-09 03:46:40.000000000 +0200 @@ -16,7 +16,7 @@ sampleW :: U.Vector (Double,Double) sampleW = U.zip sample (U.reverse sample) --- Comlex vector for FFT tests +-- Complex vector for FFT tests sampleC :: U.Vector (Complex Double) sampleC = U.zipWith (:+) sample (U.reverse sample) diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/changelog.md new/statistics-0.16.1.0/changelog.md --- old/statistics-0.16.0.1/changelog.md 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/changelog.md 2001-09-09 03:46:40.000000000 +0200 @@ -1,3 +1,14 @@ +## Changes in 0.16.1.0 + + * Dependency on monad-par is dropped. `parMap` from `parallel` is used instead. + + +## Changes in 0.16.0.2 + + * Bug in constructor of binomial distribution is fixed (#181). It accepted + out-of range probability before. + + ## Changes in 0.16.0.0 * Random number generation switched to API introduced in random-1.2 @@ -21,9 +32,9 @@ * Test suite is finally fixed (#42, #123). It took very-very-very long time but finally happened. - * Avoid loss of precision when computing CDF for exponential districution. + * Avoid loss of precision when computing CDF for exponential distribution. - * Avoid loss of precision when computing CDF for geometric districution. Add + * Avoid loss of precision when computing CDF for geometric distribution. Add complement of CDF. * Correctly handle case of n=0 in poissonCI @@ -251,17 +262,17 @@ * Accesors for uniform distribution are added. - * ContGen instances for all continuous distribtuions are added. + * ContGen instances for all continuous distributions are added. * Beta distribution is added. - * Constructor for improper gamma distribtuion is added. + * Constructor for improper gamma distribution is added. * Binomial distribution allows zero trials. * Poisson distribution now accept zero parameter. - * Integer overflow in caculation of Wilcoxon-T test is fixed. + * Integer overflow in calculation of Wilcoxon-T test is fixed. * Bug in 'ContGen' instance for normal distribution is fixed. @@ -302,7 +313,7 @@ * Root finding is added, in S.Math.RootFinding. * The complCumulative function is added to the Distribution - class in order to accurately assess probalities P(X>x) which are + class in order to accurately assess probabilities P(X>x) which are used in one-tailed tests. * A stdDev function is added to the Variance class for @@ -317,7 +328,7 @@ * Bugs in quantile estimations for chi-square and gamma distribution are fixed. - * Integer overlow in mannWhitneyUCriticalValue is fixed. It + * Integer overflow in mannWhitneyUCriticalValue is fixed. It produced incorrect critical values for moderately large samples. Something around 20 for 32-bit machines and 40 for 64-bit ones. @@ -339,18 +350,18 @@ * Mean and variance for gamma distribution are fixed. - * Much faster cumulative probablity functions for Poisson and + * Much faster cumulative probability functions for Poisson and hypergeometric distributions. * Better density functions for gamma and Poisson distributions. * Student-T, Fisher-Snedecor F-distributions and Cauchy-Lorentz - distrbution are added. + distribution are added. * The function S.Function.create is removed. Use generateM from the vector package instead. - * Function to perform approximate comparion of doubles is added to + * Function to perform approximate comparison of doubles is added to S.Function.Comparison * Regularized incomplete beta function and its inverse are added to diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/statistics.cabal new/statistics-0.16.1.0/statistics.cabal --- old/statistics-0.16.0.1/statistics.cabal 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/statistics.cabal 2001-09-09 03:46:40.000000000 +0200 @@ -1,5 +1,5 @@ name: statistics -version: 0.16.0.1 +version: 0.16.1.0 synopsis: A library of statistical types, data, and functions description: This library provides a number of common functions and types useful @@ -120,14 +120,13 @@ , binary >= 0.5.1.0 , primitive >= 0.3 , dense-linear-algebra >= 0.1 && <0.2 + , parallel >= 3.2.2.0 && <3.3 , vector >= 0.10 , vector-algorithms >= 0.4 , vector-th-unbox , vector-binary-instances >= 0.2.1 , data-default-class >= 0.1.2 - if !impl(ghcjs) - build-depends: - monad-par >= 0.3.4 + -- Older GHC if impl(ghc < 7.6) build-depends: diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/tests/Tests/Distribution.hs new/statistics-0.16.1.0/tests/Tests/Distribution.hs --- old/statistics-0.16.0.1/tests/Tests/Distribution.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/tests/Tests/Distribution.hs 2001-09-09 03:46:40.000000000 +0200 @@ -166,9 +166,9 @@ -- -- > CDF(i) - CDF(i-e) = P(i) -- - -- Apporixmate equality is tricky here. Scale is set by maximum - -- value of CDF and probability. Case when all proabilities are - -- zero should be trated specially. + -- Approximate equality is tricky here. Scale is set by maximum + -- value of CDF and probability. Case when all probabilities are + -- zero should be treated specially. badN = [ printf "N=%3i p[i]=%g\tp[i+1]=%g\tdP=%g\trelerr=%g" i p p1 dp ((p1-p-dp) / max p1 dp) | i <- [0 .. 100] , let p = cumulative d $ fromIntegral i - 1e-6 @@ -277,9 +277,9 @@ $ counterexample (printf "Sum = %g" p2) $ counterexample (printf "Delta = %g" (abs (p1 - p2))) $ abs (p1 - p2) < 3e-10 - -- Avoid too large differeneces. Otherwise there is to much to sum + -- Avoid too large differences. Otherwise there is to much to sum -- - -- Absolute difference is used guard againist precision loss when + -- Absolute difference is used guard against precision loss when -- close values of CDF are subtracted where n = min a b @@ -360,7 +360,7 @@ instance Param ExponentialDistribution instance Param GammaDistribution where -- We lose precision near `incompleteGamma 10` because of error - -- introuced by exp . logGamma. This could only be fixed in + -- introduced by exp . logGamma. This could only be fixed in -- math-function by implementing gamma prec_quantile_CDF _ = (24,24) prec_logDensity _ = 64 diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn' '--exclude=.svnignore' old/statistics-0.16.0.1/tests/Tests/Helpers.hs new/statistics-0.16.1.0/tests/Tests/Helpers.hs --- old/statistics-0.16.0.1/tests/Tests/Helpers.hs 2001-09-09 03:46:40.000000000 +0200 +++ new/statistics-0.16.1.0/tests/Tests/Helpers.hs 2001-09-09 03:46:40.000000000 +0200 @@ -60,8 +60,8 @@ -- Check that function is nondecreasing taking rounding errors into -- account. -- --- In fact funstion is allowed to decrease less than one ulp in order --- to guard againist problems with excess precision. On x86 FPU works +-- In fact function is allowed to decrease less than one ulp in order +-- to guard against problems with excess precision. On x86 FPU works -- with 80-bit numbers but doubles are 64-bit so rounding happens -- whenever values are moved from registers to memory monotonicallyIncreasesIEEE :: (Ord a, IEEE.IEEE b) => (a -> b) -> a -> a -> Bool
participants (1)
-
Source-Sync